Robotics graduate student with strong software expertise in C++/Python, ROS, and algorithm development, specializing in control algorithm and reinforcement learning for robotic systems. Experienced in building and integrating simulation platforms and intelligent decision-making pipelines. Complemented by a solid mechanical engineering background, including CAD modeling and hardware integration.
Developed a high-fidelity aerodynamic and rigid-body dynamics model for high-speed spinning cards to study trajectory prediction and stability
Built an aerodynamics + rigid dynamics simulation platform in MuJoCo, integrated with MoveIt! and a Franka Emika Panda robotic arm for throwing experiments.
Explored reinforcement learning–based trajectory optimization methods to improve throwing accuracy and consistency
Completed the simulation stage; currently progressing toward sim-to-real validation, with plans to submit a related paper to IEEE Robotics and Automation Letters (RAL).
Reach Truck Mast Vibration Control – Modeling & Control (MATLAB/Simulink)
Investigated mast vibration under acceleration and payload variation to improve operational safety and positioning accuracy.
Built the mast dynamics model using a lumped-mass and flexible-beam formulation; derived state-space representation for control design.
Designed a differential-flatness–based feedforward controller, and implemented feedback with Proportional and Sliding-Mode Control (SMC) to suppress residual oscillations.
Implemented a MATLAB/Simulink simulation and evaluated responses across multiple operating conditions.
Results show the feedforward + SMC strategy significantly reduces vibration amplitude and settling time versus a P-only baseline, validating its effectiveness for improving load stability in reach trucks.